key: cord-0779818-64kqxpju authors: Ghoshal, Uday C.; Srivastava, Deepakshi; Verma, Abhai; Ghoshal, Ujjala title: Tropical Sprue in 2014: the New Face of an Old Disease date: 2014-04-30 journal: Curr Gastroenterol Rep DOI: 10.1007/s11894-014-0391-3 sha: 28efcb420335a1d579b1fac3d4d1fc2f8841d360 doc_id: 779818 cord_uid: 64kqxpju Tropical sprue (TS), once known to be a common cause of malabsorption syndrome (MAS) in India and other tropical countries, is believed to be uncommon currently in spite of contrary evidence. Several recent studies from India showed TS to be the commonest cause of sporadic MAS in Indian adults. TS is diagnosed in patients presenting with suggestive clinical presentation, which cannot be explained by another cause of MAS and investigations revealing malabsorption of two unrelated substances, abnormal small-intestinal mucosal histology, which responds to treatment with antibiotics such as tetracycline and folic acid. There is substantial overlap between TS and postinfectious irritable bowel syndrome. There have been several advances in epidemiology, pathogenesis, and diagnosis of TS, hitherto an enigmatic condition. Tropical sprue (TS) or tropical malabsorption, believed to have disappeared even in tropical countries despite contradictory evidence, is the commonest cause of sporadic malabsorption syndrome (MAS) in northern, southern, and eastern parts of India even currently [1••, 2•, 3, 4, 5••]. It is a condition in which malabsorption occurs in absence of an identified specific etiology primarily in the tropical regions of the world [6, 7•] . TS is diagnosed based on the following criteria: (a) compatible clinical presentation, (b) demonstration of malabsorption of two unrelated substances, (c) abnormal smallintestinal mucosal histology, which may be patchy, (d) exclusion of other specific causes for MAS (except small-intestinal bacterial overgrowth [SIBO]), and (e) persistent response to treatment with antibiotics such as tetracycline and folic acid [2•] . Compatible clinical presentation is very important as abnormal small-bowel mucosal histology has been reported in tropics even in asymptomatic subjects, a condition called as tropical enteropathy [8] . Tropical enteropathy, however, is not same as TS ( Table 1) . How Common Was TS in the Past? In the past, most believed that TS was the commonest cause of MAS in Indian adults [9] [10] [11] . In fact, epidemics of chronic diarrhea and MAS following acute infective gastroenteritis (referred to as "postinfectious tropical malabsorption") were reported from India [12•] . In a study from four villages from southern India in 1968, 8-20 % villagers, suffering from an attack of acute infective diarrhea continued to experience liquidity and frequency of stool who were diagnosed to have TS [1••] . Several other studies reported epidemics of diarrhea with features suggestive of TS in India [13] [14] [15] [16] . The Madras State health reports recorded five epidemics of chronic diarrhea of unknown etiology that clinically resembled TS between 1930 and 1942 [17] . Epidemics of TS also occurred among soldiers and prisoners of war in the Indo-Burma region during the Second World War, in American military personnel serving in the Philippines and in Bangladesh [6, 7•, 18] . A similar condition was also reported sporadically from temperate countries where it was named "temperate sprue" [19] . It has been conventionally believed that with improvement in socioeconomic status and hygiene and increased use of antibiotics and probiotics, there has been a reduction in frequency of sporadic TS. There is, however, hardly any evidence in favor of this hypothesis; in contrast, there has been several studies reported recently that contradict this hypothesis. In a recent study on 275 patients with sporadic MAS from Lucknow, India, 101 (37 %) had TS, which was the single commonest cause of MAS [5••] . In another study from a large teaching hospital in New Delhi, India, though celiac disease was the single commonest cause of MAS (62 %), 22 % had TS [3] . In a study reported from a hospital in Vellore, located in the southern region of India, 36 (29 %) of 124 adult patients with sporadic MAS had TS, which was the single commonest cause in this population [4] . In an earlier study from Lucknow, India, 39/99 (39 %) patients with MAS were diagnosed having TS using very rigorous criteria [2•] . All these studies clearly contradict the belief that sporadic form of TS has disappeared currently. In fact, most reports have suggested that the sporadic form of TS is frequently seen in India even today [20•] . Patients with TS have also been reported sporadically even recently from other parts of the world. Two American subjects living in Thailand and Vietnam were reported to have TS in 2013 [21] . Two other cases who developed TS after returning from Singapore, Thailand, Indonesia, and Peru were reported in 2010 from the USA [22] . However, epidemics of TS have not been reported recently. This argument has been used to support the hypothesis that frequency of TS has reduced currently. An important issue to be reviewed here relates to a recently described entity called postinfectious irritable bowel syndrome (PI-IBS) [23•] . In this condition, following an episode of acute gastrointestinal infection associated with at least two of the followings (a) diarrhea, (b) vomiting, (c) fever, and (d) isolation of enteropathogen on stool culture, patients develops IBS for the first time [23•] . Most patients with PI-IBS have diarrhea-predominant disease [23•] . Initial studies on PI-IBS were reported from Sheffield, UK [24] [25] [26] [27] . Subsequently, several studies have been reported from other temperate countries in the world such as Canada, North America, South Korea, and China [28] [29] [30] [31] [32] . Would this condition, in which stool continued to remain frequent and liquid after an episode of acute gastroenteritis, be called as temperate sprue if described five decades ago? The answer to this question would be affirmative if the patients were found to have biochemical evidence of mucosal malabsorption on D-xylose test, fecal fat estimation, and abnormal small-intestinal histology. However, in no study that described PI-IBS, tests for mucosal malabsorption and small-intestinal mucosal histology were reported. Hence, it is difficult to say whether a proportion of these patients with socalled PI-IBS could have suffered from postinfectious tropical malabsorption or TS. Obviously, this form of postinfectious tropical malabsorption is similar to epidemic form of TS. There is considerable overlap in epidemiology, pathogenesis, clinical presentation, and treatment of postinfectious tropical malabsorption or TS and PI-IBS (Table 2) . Briefly, 7-31 % patients with acute gastroenteritis have been reported to develop PI-IBS, a frequency quite similar (8-20 %) to that reported for postinfectious tropical malabsorption, whereas the latter condition was reported about five to six decades ago, interest in the former condition emerged primarily during last two decades [7•, 33] . Both the conditions have similar pathogenesis and treatment ( Table 2) . Though several studies on PI-IBS have been reported from temperate countries recently, not a single study on this condition has been reported recently in the areas of the world where epidemic of TS was reported [23•, 34] . Unfortunately, not a single study reported till date on PI-IBS performed tests for mucosal malabsorption and small-intestinal biopsy to exclude that these patients were suffering from PI-IBS and not postinfectious tropical malabsorption. In fact, a few studies reported abnormal smallintestinal permeability (SIP) in patients with diarrheapredominant IBS (IBS-D) with or without SIBO [35, 36] , some of whom might have PI-IBS; in fact, abnormal SIP might suggest presence of mucosal malabsorption; and abnormal SIP is well known in patients with TS [37••] . Therefore, based on the above evidence, it is logical to believe that a disorder somewhat similar to that described from southern We believe that following an acute gastroenteritis, though some patients recover completely, others may continue to have bowel dysfunction, which may be due to postinfectious tropical malabsorption or due to PI-IBS ( Fig. 1 ). Tests for mucosal malabsorption and small-intestinal histology are essential to differentiate between these two conditions. Hence, we suggest that in the future iteration of Rome criteria for diagnosis of PI-IBS, it is important to consider this issue and modify the criteria for diagnosis of PI-IBS as suggested in Table 3 . Pathogenesis of TS is multifactorial [38] . These factors include, gut dysbiosis including small-intestinal bacterial colonization and overgrowth, dysregulated host's immune response leading to protracted inflammation, reduced gut defense, abnormal SIP, villous atrophy, mucosal disaccharidase deficiency including that of lactase, bile acid de-conjugation, neuro-hormonal dysregulation, exaggerated fat-induced ileal brake leading to slow mouth-to-cecum transit promoting bacterial colonization, deficiency of vitamin B 12 and folate causing megaloblastosis of small-intestinal epitheliums, and colonic dysfunction leading to reduced colonic water Bacterial colonization (10/13, 77 % vs. 3/12, 25 %, respectively) and overgrowth (4/13, 31 % vs. 0/12, 0 %, respectively) were commoner in patients with TS than controls as demonstrated in an earlier study [39••] . Bacteria in the small intestine alters enterocyte brush-border motility [42] , deconjugate bile acids preventing its enterohepatic circulation [43] . As de-conjugated bile acids cannot form chylomicrons, fatty acid absorption is impaired causing steatorrhea [44] . Deconjugated bile acids also directly inhibit carbohydrate transporters, reduce intraluminal pH, and further damage intestinal enterocytes creating a self-perpetuating cycle [42] . Bacteria produce other metabolites such as acetate and formate [44] , which may affect intestinal function. Bacterial colonization and overgrowth cause alteration of duodenal morphology, namely, villous atrophy and an increase in intra-epithelial lymphocytes, which return to normal on antibiotic treatment [39••] . In TS, many different bacteria have been isolated including Escherichia coli, Klebsiella pneumoniae, and Enterobacter cloace, and patients can harbor one or more highly toxigenic strains [45] . Bacterial colonization in the upper gut is known to occur in children and adults with acute diarrheal disease in temperate areas of the world [46, 47] . Similar findings have been shown in tropical areas such as in Calcutta (currently renamed Kolkata) in association with acute gastroenteritis due to E. coli, Shigella, Vibrio cholerae, or a variety of nonspecific enteric bacteria [48] . Moreover, bacterial contamination in the jejunum was found to persist in these patients up to 4 months after the original infection [48] . One study reported reduced secretion of gastric acid in patients with TS, though salivary IgA was normal [49] . In another study, a proportion of patients with TS were found to have hypogammaglobulinemia [50] . Obviously, these factors would result into reduced innate immunity of gut causing bacterial colonization and overgrowth in the small bowel in the presence of inadequate food hygiene, which is not uncommon in areas of the world where TS has been reported. Zinc supplementation is known to improve gut defense and improve SIP. Zinc supplementation reduced the severity and duration of diarrhea in infants and young children, in an earlier study from India [51] . But, there is no study of zinc in relation to TS. In a report on a single patient with TS developing following journey to Sri Lanka, low level of zinc was detected [52] . However, more studies using sophisticated immunological techniques may uncover many other aspects of reduced gut defense in patients with TS. Two earlier studies suggested that protozoa and fungi may play an important role in the pathogenesis of TS [53, 54] . Cryptosporidium parvum, Isospora belli, Blastocystis hominis, and Cyclospora cayetanensis have been identified in fecal samples and small-intestinal biopsy specimens of people suffering from various forms of prolonged diarrhea in the tropics [53] [54] [55] . Moreover, viral particles resembling orthomyxo and coronaviruses have been demonstrated in the feces of 90 % of patients with TS from southern India [56] . Normal SIP is necessary for physiological function of the small bowel including absorption [57] . SIP is abnormal in several diseases associated with altered small-intestinal functions such as IBS-D, PI-IBS, inflammatory bowel disease, and celiac disease [58] [59] [60] [61] [62] . Postdysenteric gut dysfunction might result from persistent low-grade inflammation in enteric submucosal and neuromuscular compartments, as shown from studies in animals and human [63] . In patients with IBS-D and PI-IBS, increase in permeability could promote neuromuscular inflammation that disturbs enteric sensation and motility. Abnormal SIP has been reported in patients with TS. A study on estimation of SIP using urinary excretion of lactulose and mannitol by 1H-NMR spectroscopy on a small number of patients with MAS (n=18), including eight patients with TS and healthy controls (n=28), showed that the urinary excretion ratio of lactulose to mannitol was higher in patients with MAS as compared to healthy controls [64] . Also, urinary excretion of lactulose was higher in patients with TS but excretion of mannitol was the same as compared with healthy controls [64] . Another study on SIP using urinary excretion of lactulose and L-rhamnose showed that urinary excretion of lactulose was higher in TS patients as compared with healthy controls [65] . The magnitude of the absorption defects demonstrated in patients with TS was more severe than would be anticipated from the jejunal mucosal abnormalities [65] . In another study, abnormal SIP among patients with TS persisted in spite of treatment in a proportion of patients; [37••] those with persistently abnormal SIP continued to pass more frequent stool and gained less weight than those in whom it got normalized [37••] . Is TS Related to Neuro-Hormonal Dysregulation? TS is known to be associated with small-bowel stasis, which predispose to small-intestinal bacterial colonization and overgrowth [39••] . Small-bowel stasis in patients with TS may result from exaggerated ileal brake induced by malabsorbed fat passing through the ileum, which liberate gut hormones like peptide YY, neurotensin, and glucagon-like peptide-1 (Fig. 2) [39••, 66••]. Gut hormones play a major role in intestinal motility. Hormones such as enteroglucagon, peptide YY, and neurotensin slow transit while motilin enhances the rate of gastric emptying and hastens small-intestinal motility [67, 68] . In an earlier study, patients with TS had raised basal plasma motilin and enteroglucagon concentrations, but their postprandial release of both gastric inhibitory polypeptide and insulin was significantly reduced [69] . In two earlier studies, orocaecal transit time (OCTT) was found to be longer in patients with TS than in controls [ [70] . In a recent study, infusion of fat into the proximal small intestine was shown to inhibit antroduodenal motility, delay duodeno-cecal transit time, and increase plasma levels of peptide YY and neurotensin than among controls [66••] . Though the same mechanism of fatinduced ileal brake may operate in patients with MAS and steatorrhoea due to any cause, such as celiac disease and chronic pancreatitis, it has been studied only in patients with TS [66••]. In patients with MAS such as TS, a larger amount of fluid passes through the ileocecal valve than the healthy subjects [66••] . Colon can increase its capacity to absorb electrolytes and water to compensate for the higher amount of fluid coming to the colon to prevent occurrence of diarrhea. Ramakrishna et al. showed that in patients with TS, there is abnormality in the colon that prevents this compensatory mechanism predisposing to occurrence of diarrhea [71] [72] [73] . Is TS Due to Vitamin B 12 and Folate Malaborption? Deficiencies of folate and vitamin B 12 may favor the development of TS by causing megaloblastosis of small-intestinal epithelium [74] . Vegetarians are at risk for vitamin B 12 deficiency due to suboptimal intake [75] . A large proportion of Indian subjects are vegetarian, which may predispose to the deficiency of vitamin B 12 . Moreover, SIBO may predispose to vitamin B 12 deficiency as occurring in blind loop syndrome [76] . Although vitamin B 12 absorption is frequently improved by antibiotics, it may not return to normal levels despite prolonged therapy [77] . There may be, therefore, a coexistent mucosal lesion as previously described in E. coli-contaminated rats and small-intestinal diverticulosis in man [78, 79] . In a previous study on 38 patients with vitamin B 12 malabsorption, antibiotic therapy resulted in a rapid return of vitamin B 12 absorption to normal in 50 % of the patients [12•] . In another study, vitamin B 12 absorption returned to normal and there was a marked decrease in total luminal flora in all three subjects who were treated with tetracycline [79] . Furthermore, there was some correlation between the bacteriological findings in both the jejunum and ileum and vitamin B 12 malabsorption [80] . In the jejunal aspirates of patients with vitamin B 12 malabsorption, the mean log counts of aerobic flora, total flora, and bacteroides were greater than the counts in the controls and in the ileal aspirates, all except one of the patients with vitamin B 12 malabsorption had bacteria present in counts greater than 10 4 colony forming units/milliliter (CFU/mL) [80] . Whereas in those with normal vitamin B 12 absorption, only one fourth had bacterial growth and none had more than 10 3 CFU/mL [80] . This might suggest that bacteria to produce toxins altering small-intestinal physiology. Such a mechanism has been postulated for E. coli in a case of diverticulosis [79] . However, further studies to examine the nature of the cellular lesion responsible for the vitamin B 12 absorptive defect in TS and the possible role of bacteria in its pathogenesis are required. Empirical antibiotic treatment has been shown to improve diarrhea and associated malnutrition in most patients with TS [77, 81] . Most studies have been done with tetracycline and sulfonamides [81, 82] . It has been shown that folate and vitamin B 12 absorption improves markedly and symptoms resolve early with antibiotics [83] . Treatment with folic acid alone can be curative early in the course of the disease [84] . It has been proposed that antibiotics reduce the overall bacterial load in the gut leading to reduced inflammation and thus improving diarrhea [85] . Prolonged treatment with tetracycline may be required for complete resolution of symptoms. Tetracycline is generally given for 2 months. In a recent study, rifaximin was tried in treatment of TS without encouraging result [86] . More studies, however, are needed on this issue. Adequate dietary management is the key for the management of TS. Steatorrhea is a common feature of TS and contributes significantly to diarrhea [87, 88] . Antibiotics and vitamins may not completely cure steatorrhea. Cause of steatorrhea is multifactorial-mucosal malabsorption of fat, bacteria-splitting bile acids, and impaired enterohepatic recirculation of bile salts [89] . Restriction of long-chain fatty acids may help in reducing steatorrhea. It should be replaced by medium-chain triglycerides. Coconut oil, which is rich in mediumchain fatty acids, may be helpful in reducing diarrhea [90] . Moreover, malabsorbed fat might cause exaggerated ileal brake reducing gut motility, promoting SIBO and malabsorption in TS [39••] . Protein and carbohydrate intake should be guided by the weight and body mass index. Long-term prognosis of patients with TS with proper dietary management and antibiotics plus vitamins is good. However, in endemic regions, the disease may have relapsing course requiring repeated cycles of antibiotics. Also, not all patients who improve symptomatically have complete resolution of small-intestinal histology and functions such as SIP [39••, 91] . TS, once thought to be a common cause of MAS, is now believed to be uncommon, though the evidence contradicts this belief. There is significant overlap between TS and PI-IBS. In fact, PI-IBS and TS may be continuing spectrum of response of the gut to an episode of infectious gastroenteritis (Fig. 4) . Since pathogenesis of TS and PI-IBS is similar, we may need to learn about management of PI-IBS from our knowledge on management of TS. More research is needed on this issue. Epidemic tropical sprue and other epidemics of diarrhea in South Indian villages Etiological spectrum of sporadic malabsorption syndrome in northern Indian adults at a tertiary hospital Current spectrum of malabsorption syndrome in adults in India Spectrum of malabsorption in India-tropical sprue is still the leader Spectrum of malabsorption syndrome among adults and factors differentiating celiac disease and tropical malabsorption Outbreaks of sprue during the Burma Campaign A perspective on tropical sprue Environmental enteropathy: critical implications of a poorly understood condition Tropical sprue Bacterial flora of the gastrointestinal tract in southern Indian control subjects and patients with tropical sprue Tropical sprue and malnutrition in West Bengal. II. Fluid and electrolyte transport in the small intestine Syndrome of tropical sprue in South India Particulars regarding the prevalence of diarrhea at Simla in 1880 Sprue in India Dietetic deficiency syndromes in Indian soldiers Clinical features and pathogenesis of tropical sprue; observations on a series of cases among Italian prisoners of war in India Trop med couf roy coil physicians. London: Pitman. I Digestive disease as a national problem. VI. Enteric disease among United States troops in Vietnam Temperate sprue Is the ghost of tropical sprue re-surfacing after its obituary? Tropical sprue: two cases report and review of literature. Has anything changed in the past 40 years The face of tropical sprue in 2010 Post-infectious irritable bowel syndrome: the past, the present and the future Psychometric scores and persistence of irritable bowel after infectious diarrhoea Irritable bowel syndrome-post salmonella infection Prevalence of gastrointestinal symptoms six months after bacterial gastroenteritis and risk factors for development of the irritable bowel syndrome: postal survey of patients The role of psychological and biological factors in postinfective gut dysfunction Post-traveler's diarrhea irritable bowel syndrome: a prospective study The incidence of irritable bowel syndrome among community subjects with previous acute enteric infection Bacillary dysentery as a causative factor of irritable bowel syndrome and its pathogenesis Postdiarrhea chronic intestinal symptoms and irritable bowel syndrome in North American travelers to Mexico Post-infectious irritable bowel syndrome in patients with Shigella infection Post-infectious irritable bowel syndrome Bugs and irritable bowel syndrome: the good, the bad and the ugly The relationship between small-intestinal bacterial overgrowth and intestinal permeability in patients with irritable bowel syndrome Abnormal intestinal permeability in subgroups of diarrheapredominant irritable bowel syndromes Abnormal small intestinal permeability in patients with idiopathic malabsorption in tropics (tropical sprue) does not change even after successful treatment Tropical malabsorption Tropical sprue is associated with contamination of small bowel with aerobic bacteria and reversible prolongation of orocecal transit time Duodenal mucosal morphometry of elderly patients with small intestinal bacterial overgrowth: response to antibiotic treatment What is tropical sprue? Review: pathophysiology of diarrhea caused by bacterial overgrowth of the small intestine Bacterial populations contaminating the upper gut in patients with small intestinal bacterial overgrowth syndrome Malabsorption syndrome with and without small intestinal bacterial overgrowth: a study on upper-gut aspirate using 1H NMR spectroscopy Enterotoxigenicity of colonising coliform bacteria in tropical sprue and blind-loop syndrome The role of certain varieties of bacterium coli in gastroenteritis of babies Microbial intestinal flora in acute diarrheal disease Acute undifferentiated human diarrhea in the tropics. I. Alterations in intestinal micrflora Salivary and gastric secretions in patients with tropical sprue Dysgammaglobulinaemia in tropical sprue Zinc supplementation in young children with acute diarrhea in India Chronic diarrhea after a journey to Sri Lanka Epidemiology of diarrhoeal illness associated with coccidian-like organism among travellers and foreign residents in Nepal An alga-like organism associated with an outbreak of prolonged diarrhea among foreigners in Nepal Aetiology and pathogenesis of postinfective tropical malabsorption (tropical sprue) Pleomorphic virus-like particles in human faeces Intestinal permeability tests and integrity of the small intestinal mucosa Intestinal membrane permeability and hypersensitivity in the irritable bowel syndrome Increased rectal mucosal enteroendocrine cells, T lymphocytes, and increased gut permeability following acute Campylobacter enteritis and in post-dysenteric irritable bowel syndrome Intestinal permeability in patients with irritable bowel syndrome after a waterborne outbreak of acute gastroenteritis in Walkerton Abnormal small intestinal permeability to sugars in patients with Crohn's disease of the terminal ileum and colon In vitro determination of small intestinal permeability: demonstration of a persistent defect in patients with coeliac disease Gut permeability in irritable bowel syndrome: more leaks add to slightly inflamed bowel syndrome conspiracy theory Assessment of small intestinal permeability using 1H-NMR spectroscopy Intestinal absorption and unmediated permeation of sugars in post-infective tropical malabsorption (tropical sprue) Pathogenesis of tropical sprue: a pilot study of antroduodenal manometry, duodenocaecal transit time and fat-induced ileal brake Effect of motilin on the rate of gastric emptying and gut hormone release during breakfast Gut hormones in tropical malabsorption Utility of hydrogen breath tests in diagnosis of small intestinal bacterial overgrowth in malabsorption syndrome and its relationship with oro-cecal transit time Water and electrolyte absorption by the colon in tropical sprue Role of bacterial toxins, bile acids, and free fatty acids in colonic water malabsorption in tropical sprue Absorption of water and sodium and activity of adenosine triphosphatases in the rectal mucosa in tropical sprue Tropical sprue How prevalent is vitamin B(12) deficiency among vegetarians? Transient selective vitamin B12 malabsorption in an adult Long-term antibiotic therapy in tropical sprue The effect of E. coli on the absorption of vitamin B12 Diverticulosis of the small intestine and megaloblastic anemia: intestinal microflora and absorption before and after tetracycline administration Vitamin B12 absorptiona study of intraluminal events in control subjects and patients with tropical sprue Long-term follow-up of antibiotic-treated tropical sprue Response of tropical sprue to poorly absorbed sulfonamides and oxytetracycline Assessment of early and delayed responses in vitamin B12 absorption during antibiotic therapy in tropical malabsorption Jejunal histology and clinical status in tropical sprue and other chronic diarrheal disorders Small intestinal bacterial overgrowth: a comprehensive review A randomized, double-blind, placebo-controlled trial of rifaximin, a nonabsorbable antibiotic, in the treatment of tropical enteropathy Idiopathic tropical diarrhea with or without steatorrhea (tropical malabsorption syndrome) Steatorrhea in tropical sprue patients in Puerto Rico Diffuse primary fat replacement of the pancreas: an unusual cause of steatorrhea Antimicrobial effects of virgin coconut oil and its mediumchain fatty acids on Clostridium difficile Abnormal small intestinal permeability in patients with idiopathic malabsorption in tropics (tropical sprue) does not change even after successful treatment Conflict of Interest Uday C. Ghoshal, Ujjala Ghoshal, Deepakshi Srivastava, and Abhai Verma declare no conflict of interest. This article does not contain any studies with human or animal subjects performed by the author.